Optimal Resource Allocation for U-Shaped Parallel Split Learning
Authors:
- 4 December 2023
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 197-202
Abstract
Split learning (SL) has emerged as a promising approach for model training without revealing the raw data samples from the data owners. However, traditional SL inevitably leaks label privacy as the tail model (with the last layers) should be placed on the server. To overcome this limitation, one promising solution is to utilize U-shaped architecture to leave both early layers and last layers on the user side. In this paper, we develop a novel parallel U-shaped split learning and devise the optimal resource optimization scheme to improve the performance of edge networks. In the proposed framework, multiple users communicate with an edge server for SL. We analyze the end-to-end delay of each client during the training process and design an efficient resource allocation algorithm, called LSCRA, which finds the optimal computing resource allocation and split layers. Our experimental results show the effectiveness of LSCRA and that U-shaped parallel split learning can achieve a similar performance with other SL baselines while preserving label privacy.Keywords
All Related Versions
This publication has 17 references indexed in Scilit:
- Vehicle as a Service (VaaS): Leverage Vehicles to Build Service Networks and Capabilities for Smart CitiesIEEE Communications Surveys & Tutorials, 2024
- Channel Power Gain Estimation for Terahertz Vehicle-to-Infrastructure NetworksIEEE Communications Letters, 2022
- Edge Intelligence for Mission-Critical 6G Services in Space-Air-Ground Integrated NetworksIEEE Network, 2022
- LEOPARD: Parallel Optimal Deep Echo State Network Prediction Improves Service Coverage for UAV-Assisted Outdoor HotspotsIEEE Transactions on Cognitive Communications and Networking, 2021
- Machine-Learning-Aided Mission-Critical Internet of Underwater ThingsIEEE Network, 2021
- A Survey on Federated Learning for Resource-Constrained IoT DevicesIEEE Internet of Things Journal, 2021
- Privacy-Sensitive Parallel Split Learning2020
- The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesionsScientific Data, 2018
- Distributed learning of deep neural network over multiple agentsJournal of Network and Computer Applications, 2018
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998